Monitoring & Notifications on Unraid

Why Monitoring Matters

Hard drives rarely die all at once with no warning. Far more often they give little signals for days or weeks first - a rising error count, a strange temperature, a slowdown. Monitoring is simply catching those signals.

The whole idea is to move from surprised to prepared. Without monitoring, the first you hear of a problem is when a drive is already dead and files are missing. With monitoring, you hear about it while the drive is still working, when swapping it is calm and routine. Same problem, completely different day.

• Early warning - you learn about trouble while there is still time to act calmly.
• Confidence - no news really is good news, because you know you would be told.
• Fewer surprises - problems announce themselves instead of ambushing you.

Turning On Notifications in Settings

Everything starts here. Unraid has a notifications page in its settings where you switch on the whole system and decide how often it checks in.

Notifications come in levels, usually something like normal, warning, and alert. Warnings and alerts are the ones you care about - they mean something needs your attention. You can let the gentle informational notices show on-screen only, and have the serious ones actually reach your phone or inbox.

• Open Unraid in your web browser as you normally do.
• Go to the Settings area and find the Notifications page.
• Turn notifications on so the server begins watching and reporting.
• Choose how often it checks - a regular background check is fine.

The Ways Alerts Can Reach You

A notification is only useful if it finds you. Unraid can deliver alerts several ways, and you can use more than one at the same time for good measure.

For most people, email plus a phone push app is the sweet spot. Email gives you a searchable record, and a push alert reaches you quickly when you are away from the computer. Set up at least one that reaches you off the server itself.

Drive-Health (SMART) Alerts

Every modern hard drive keeps a little health diary about itself. This is called SMART. Unraid can read that diary and warn you the moment a drive starts logging trouble.

SMART tracks things like the number of bad spots a drive has had to work around, and how many read or write errors it has hit. A healthy drive holds these steady near zero. A failing drive shows them climbing. Unraid watches for that climb and raises a flag - which is your cue to plan a swap.

• It warns you while the drive still works, so nothing is lost yet.
• It turns a scary drive death into a planned, calm replacement.
• In an array, catching one weak drive early protects all the rest.

Temperature Alerts for Drives and CPU

Heat is the quiet enemy of any computer. Drives and processors both have comfortable temperature ranges, and running hot for a long time shortens their lives. Unraid can warn you before things get too warm.

You do not need to memorize exact numbers. Unraid comes with reasonable defaults, and you can nudge them. The real value is the alert itself: if a fan dies or a vent clogs with dust, a temperature warning tells you long before the heat does any harm.

• Drives - most hard drives are happy running cool to warm. Set a warning if one climbs unusually high for your setup.
• CPU - the processor has its own comfort range. A warning here catches a stuck fan or blocked vent early.

Array and Parity Warnings

The array is the group of drives working together, and parity is the extra drive that lets Unraid rebuild a failed one. Warnings here are about the health of that whole protective system.

Unraid watches whether every drive in the array is present and behaving. If a drive drops out or gets marked as failed, you want to know right away, because that is when parity is doing its job and you should act. A parity check finding errors is another signal worth an alert.

• A disabled drive - a drive Unraid has taken offline because it misbehaved.
• Parity errors - the check found mismatches it had to consider.
• A drive missing at start - the array did not see a drive it expected.

Reading the Dashboard at a Glance

Unraid's Dashboard is your one-screen health check. Everything important is there in a glance - the array, each drive, temperatures, and how busy the server is - without digging through menus.

Green and quiet is what you want. A colored icon or a warning triangle next to a drive is Unraid pointing at exactly what needs a look. Learning to scan this page for ten seconds is one of the most useful habits you can build.

• Array status - is everything started and green, with no drive flagged?
• Each drive - a healthy temperature and no warning icon beside it.
• Space used - how full the array is, so nothing sneaks up on you.
• Activity - whether the server is calm or working hard right now.

The Fix Common Problems Plugin

This free add-on is like a friendly helper that looks over your server and points out anything set up in a risky or unusual way. It is one of the first things I install on any Unraid box.

You add it through the Community Applications plugin - the app store for Unraid. Once installed, it scans your system and shows a plain-language list of anything worth attention, from a setting that could cause trouble to a reminder you have not turned on backups. It explains each finding in words a person can actually read.

• It catches beginner mistakes before they bite you.
• It explains problems in plain English, not error codes.
• It nudges you toward the safer, recommended settings.

Disk-Space and Utilization Alerts

A drive that fills up completely can cause odd problems - failed writes, apps that misbehave, a stuck Mover. The fix is to be warned while there is still comfortable room to spare.

Unraid lets you set a warning level and an alert level for how full your disks get - for example, a gentle warning at 70 percent and a firmer alert at 90 percent. When a disk crosses that line, you hear about it, giving you time to clean up, add a drive, or move things around before it becomes urgent.

• Warning - an early heads-up, so you can plan without rushing.
• Alert - a firmer nudge that it really is time to act.
• The cache pool - watch this too, since it is often smaller and fills faster.

A Glance at the System Log

The system log is the server's running diary - a scrolling list of everything it has been doing. You do not need to read it daily, but knowing it exists helps when you are chasing down a problem.

Most of the log is routine chatter you can ignore. What you are looking for, when something is wrong, are lines that stand out - words like error or warning, often around the time the trouble started. Those lines are the clues that explain what happened, in more detail than a notification can.

• After a notification, to see the fuller story behind it.
• When a drive or app is acting up and you want the details.
• When you call me - reading me the recent errors speeds things up.

Keeping an Eye on Docker App Health

Most apps on Unraid run as Docker containers - small, self-contained programs. The Docker page shows each one and whether it is running happily, so a stopped app does not go unnoticed.

On the Docker page, each app shows its status. A running app is doing its job. One that has stopped, or keeps restarting itself, is telling you something is off - maybe an update went sideways or a setting needs a look. Catching a misbehaving app early saves you wondering later why a service quietly stopped working.

• Stopped - an app that should be running but is not.
• Restart loop - an app that keeps stopping and starting, a sign of a problem.
• Updates - Unraid flags when an app has a newer version available.

Parity-Check Reminders

A parity check is Unraid's routine self-exam - it re-reads the whole array to confirm the parity drive still matches the data, so it is ready to rebuild a failed drive if needed. Running it on a schedule keeps that safety net trustworthy.

You can set Unraid to run a parity check automatically every so often, such as once a month. It works quietly in the background, and when it finishes it reports whether everything lined up. That report is your confirmation that your protection is genuinely working, not just assumed to be.

• Schedule a regular check so it happens without you remembering.
• Let it run at a quiet time, since it makes the server busy for a while.
• Read the result - zero errors is the clean bill of health you want.

Reading a Couple of SMART Values, Simply

You do not need to be an engineer to glance at a drive's SMART report. A couple of values tell most of the story, and you are really just watching whether they stay put or start climbing.

The exact numbers matter less than the trend. Zero and steady is a happy drive. A count that was zero last month and is now growing is a drive quietly telling you it is wearing out. That is your signal to back up and plan a swap.

• Reallocated sectors - the count of bad spots the drive has had to route around. Should stay at or near zero. A rising number is a real warning sign.
• Pending sectors - spots the drive is unsure about and watching. Again, near zero is healthy; a climbing count is not.

Which Warnings to Act On Fast

Not every notification is an emergency. Knowing which ones to jump on and which can wait keeps you calm and focused on what truly protects your data.

The pattern is simple: anything about a drive's health or the array's integrity gets your attention now. Anything that is just informational, or already resolved itself, can wait for your next quiet moment at the server.

Taming False Alarms

If your server cries wolf too often, you stop listening - and that is dangerous. A little tuning keeps the alerts you get meaningful, so a real one never gets lost in the noise.

Sometimes a default threshold is set a touch too sensitive for your particular setup. A drive that always runs a bit warm, or a disk that lives comfortably at a high but stable fullness, can trip a warning that is not really a problem. Nudging those thresholds to fit your reality cuts the noise.

• Adjust a threshold that fires constantly for a condition you know is normal.
• Do not silence a category entirely - just widen it to sensible limits.
• Never turn off drive-health alerts to stop nagging - fix the drive instead.

Testing That Notifications Actually Arrive

A notification system you have never tested is just a hope. Send yourself a test and confirm it lands, so you know the pipe is clear before you ever need it in earnest.

If the test does not show up, the setup is not finished - better to learn that now than during a real drive failure. Check the address you entered, and peek in the spam folder, which is the usual culprit for a missing email alert.

• Set up your delivery method - email or a phone push app - in the settings.
• Look for a test or send test notification button on the notifications page.
• Send the test and then check that inbox or phone.
• Confirm it arrived, and that it did not land in a spam or junk folder.

A Simple Monitoring Routine

Monitoring works best as a light habit, not a chore. A few quick check-ins on a simple schedule keep your server healthy without ever eating your time.

That is genuinely the whole routine. Most of the work is done by the server itself, sending you a note only when it matters. Your part is a ten-second glance now and then, and a quick response when something pings you.

• Whenever an alert arrives - read it, decide if it is act-now or can-wait.
• Weekly - glance at the Dashboard for all-green drives and comfortable space.
• Monthly - let the scheduled parity check run and read its result.
• Yearly - a slightly closer look, and a dusting of the vents and fans.

Your Monitoring Checklist

Here is the whole job in one tidy list. Work down it once, and your server will keep watch and speak up when it needs you.

• Turn on notifications in Unraid's Settings.
• Set up email or a phone push app so alerts reach you off the server.
• Confirm drive-health (SMART) alerts are on.
• Set temperature warnings for drives and the CPU.
• Turn on disk-space warnings with sensible warning and alert levels.
• Install the Fix Common Problems plugin and read its findings.

What's in the 27-page PDF

• Why monitoring is the quiet hero of a healthy server
• Turning on notifications and choosing how they reach you
• Drive-health, temperature, and array warnings
• Reading the Dashboard and the system log at a glance
• The Fix Common Problems plugin and taming false alarms
• A simple routine and a checklist to make it stick

Why I wrote this

I write these as printable handouts because a guide you can keep in a drawer beats a web page you have to find again. No jargon, no judgment, no rushing. If you'd rather have someone sit beside you and go through it together, that's a flat $99 first visit with the 30-day follow-up included — serving Portage County from Atwater, Ohio. — Bill